AUTOMATIC CLASSIFICATION OF WALNUT LEAF IMAGES WITH GRADCAM AND DEEP LEARNING

Abstract

Walnut leaves similar color and formation make distinguishing between varieties considerably challenging for individuals. Examining and categorizing such plant leaves one by one can be a time-consuming and costly process. Hence, experimental studies are conducted in laboratory settings to classify walnut varieties. Within the scope of this study, an original dataset consisting of 1751 walnut leaf images obtained from 18 different walnut varieties was prepared. Various preprocessing techniques were applied to the original dataset, and additionally, data augmentation methods were employed to obtain an expanded dataset. Both datasets were trained using deep learning models. Among these models, the Vgg16 CNN model demonstrated the most superior performance. The proposed model, trained with Vgg16 on the augmented dataset, produced Gradcam images and was further classified using the Vgg16 CNN algorithm. According to experimental test results, the proposed model achieved a success rate of 77.11%. This study demonstrates the successful utilization of deep learning techniques for classifying walnut varieties from walnut leaf images.

Keywords

• Walnut dataset
• GradCam
• Deep learning
• Classification.

References

1. [1] K. Kulkarni, Z. Zhang, L. Chang, J. Yang, P. C. A. da Fonseca, and D. Barford, “Building a pseudo-atomic model of the anaphase-promoting complex,” Acta Crystallogr. Sect. D Biol. Crystallogr., vol. 69, no. 11, pp. 2236–2243, 2013.
2. [2] F. Doğan and İ. Türkoğlu, “Derin öğrenme algoritmalarının yaprak sınıflandırma başarımlarının karşılaştırılması,” Sak. Univ. J. Comput. Inf. Sci., vol. 1, no. 1, pp. 10–21, 2018.
3. [3] T. Karadeniz and E. Güler, “Cumhuriyetin İlk Yillarindan Günümüze Ceviz Yetiştiriciliği,” Bahçe, 2017.
4. [4] T. Karadeniz, “Ordu Yöresinde yetiştirilen ceviz genotiplerinin (Juglans regia L.) seleksiyonu,” Ordu Üniversitesi Bilim ve Teknol. Derg., vol. 1, no. 1, pp. 65–74, 2011.
5. [5] S. Solak and U. Altinişik, “Görüntü işleme teknikleri ve kümeleme yöntemleri kullanılarak fındık meyvesinin tespit ve sınıflandırılması,” Sak. Univ. J. Sci., vol. 22, no. 1, pp. 56–65, 2018.
6. [6] Y. LeCun, Y. Bengio, and G. Hinton, “Deep learning,” Nature, vol. 521, no. 7553, pp. 436–444, 2015.
7. [7] M. H. Saleem, J. Potgieter, and K. M. Arif, “Plant disease detection and classification by deep learning,” Plants, vol. 8, no. 11, p. 468, 2019.
8. [8] T. Karahan and V. Nabiyev, “Plant identification with convolutional neural networks and transfer learning,” Pamukkale Üniversitesi Mühendislik Bilim. Derg., vol. 27, no. 5, pp. 638–645, 2021.

9. [9] I. M. Dheir, A. Soliman, A. Mettleq, and A. A. Elsharif, “Nuts Types Classification Using Deep learning,” Int. J. Acad. Inf. Syst. Res., vol. 3, no. 12, pp. 12–17, 2019.
10. [10] Y. Liu, J. Su, G. Xu, Y. Fang, F. Liu, and B. Su, “Identification of grapevine (vitis vinifera l.) cultivars by vine leaf image via deep learning and mobile devices,” 2020.
11. [11] D. K. Nkemelu, D. Omeiza, and N. Lubalo, “Deep convolutional neural network for plant seedlings classification,” arXiv Prepr. arXiv1811.08404, 2018.
12. [12] A. T. Karadeniz, Y. Çelik, and E. Başaran, “Classification of walnut varieties obtained from walnut leaf images by the recommended residual block based CNN model,” Eur. Food Res. Technol., pp. 1–12, 2022.
13. [13] A. T. Karadeniz, E. Başaran, and Y. Celik, “Identification Of Walnut Variety From The Leaves Using Deep Learning Algorithms,” Bitlis Eren Üniversitesi Fen Bilim. Derg., vol. 12, no. 2, pp. 531–543, 2023.
14. [14] A. Beikmohammadi, K. Faez, and A. Motallebi, “SWP-LeafNET: A novel multistage approach for plant leaf identification based on deep CNN,” Expert Syst. Appl., vol. 202, p. 117470, 2022.
15. [15] A. Dobrescu, M. V. Giuffrida, and S. A. Tsaftaris, “Doing more with less: a multitask deep learning approach in plant phenotyping,” Front. Plant Sci., vol. 11, p. 141, 2020.
16. [16] R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra, “Grad-cam: Visual explanations from deep networks via gradient-based localization,” in Proceedings of the IEEE international conference on computer vision, 2017, pp. 618–626.
17. [17] H. Jiang et al., “A multi-label deep learning model with interpretable grad-CAM for diabetic retinopathy classification,” in 2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC), 2020, pp. 1560–1563.
18. [18] Y. Zhang, D. Hong, D. McClement, O. Oladosu, G. Pridham, and G. Slaney, “Grad-CAM helps interpret the deep learning models trained to classify multiple sclerosis types using clinical brain magnetic resonance imaging,” J. Neurosci. Methods, vol. 353, p. 109098, 2021.
19. [19] K. Simonyan and A. Zisserman, “Very deep convolutional networks for large-scale image recognition,” arXiv Prepr. arXiv1409.1556, 2014.